InMemoryProvider`, `FileMemoryProvider`,
`RedisMemoryProvider` (lazy-imported; needs the `redis` extra).
- `src/deepcrew/content.py` — multimodal input: `image()`, `pdf()`, `user_message()` build OpenAI
content blocks; `extract_text()` is the canonical way to pull plain text
dead just because a naive grep for one quote style misses it.
- Uploading a `.pdf` or `.docx` today does not extract real text — see `AUDIT.md` C1. If you're debugging
Inbox drop
Ask: "Drop 3-5 real artifacts into the `inbox/` folder — a LinkedIn PDF export, a recent Google Doc you exported, a meeting transcript, a weekly review, anything
servers the agents rely on: `context7` (library docs), `figma`, `playwright`,
`sequential-thinking`, `atlassian` (Jira), `pdf-reader`, plus AWS and GCP servers for the infra/cost commands.
Several require credentials or local
específica deste repositório (além da lista padrão do protocolo)
- <ex.: src/shared/ — usado por billing, PDF e fila>
- <ex.: src/core/tenant/ — isolamento multi-tenant>
- <ex.: src/integracoes/ — PJe, pagamento, WhatsApp>
## Multi-tenant
menos um arquivo de apoio
3. Adicionar linha na tabela do `README.md`
4. Adicionar PDF de referência em `references/books/` com padrão `Título - Autor.pdf
saúde mental (PHQ-9, GAD-7, ASRS-18) com devolutiva por IA e relatório PDF. Motor de aquisição do lançamento — SEO do lado paciente, QR no PDF do lado médico
validation" section for why a plain substring
search was replaced (it both misses compressed PDF manifests entirely
and false-positives on coincidental text).
- **Never reparse-and-reserialize a markup format
FastAPI/uvicorn API server
pip install -e ".[graphql]" # Strawberry GraphQL layer
pip install -e ".[pdf]" # PDF report export (reportlab)
# Verify installation
elengenix doctor
```
### Testing
```bash
# Full test suite (3,150+ tests
Resume Rendering (LaTeX)
Resumes and cover letters are written in **LaTeX** and compiled to PDF with `pdflatex`.
- Template: `resumes/master.tex` — uses `{{placeholders}}` that the CLI injects with real data.
- Generated files
contains the autoresearch training pipeline:
- `../corpus/` — Training corpus (parquet, txt files)
- `../build_corpus.py` — Corpus builder (PDF, web, YouTube, OCR)
- `../autoresearch-win-rtx/` — Model training pipeline (prepare.py, train.py)
- Trained model: 50.3M params
Save raw** to `raw/YYYY-MM-DD-slug.md` with metadata frontmatter:
```yaml
---
type: raw
source: url | pdf | image | text
source_url: https://...
ingested: YYYY-MM-DD
compiled_to: []
tags: []
---
```
3. **Compile** into structured wiki
research papers in Python. It defines a workflow that prioritizes local TeX sources over PDF extraction and enforces modular code design with bilingual documentation.
**Compatible with:** Codex, Claude Code, OpenClaw
A file Claude Code reads at the start of every session. It holds the commands, conventions and warnings the agent needs for this project.
Where does it go?
At the repository root. Claude Code also reads CLAUDE.md files in subdirectories when it works there.
What should it contain?
Build and test commands, the project's layout, conventions that aren't obvious from the code, and mistakes to avoid. Short files tend to work better than long ones.
CLAUDE.md or AGENTS.md?
Claude Code reads CLAUDE.md; most other agents read AGENTS.md. Many projects keep one and point the other at it.